This invention relates to novel bicyclic tetrahydroxylated pyrrolizidines and methods for their chemical synthesis. These compounds are useful inhibitors of glycosidase enzymes.
Several naturally occurring polyhydroxylated pyrrolidines, pyrrolizidines and indolizidines are powerful and specific inhibitors of glycosidases [Fellows and Fleet, Alkaloidal Glycosidase Inhibitors from Plants, in Natural Products Isolation (Ed. G. H. Wagman and R. Cooper), Elsevier, Amsterdam, 1988, pp. 540-560; Evans et al., Phytochemistry, 24 1953-1956 (1985)]. In recent years, plagiarism of plant chemistry has led to the synthesis of powerful inhibitors of other glycosidases [Fleet et al., J. Chem. Soc., Perkin Trans. 1, 665-666, (1989); Bashyal et al., Tetrahydron 3, 3083-3093 (1987), and Fleet et al., Tetrahedron 43, 979-990 (1987)]. It is now clear that, although changes in stereochemistry of the hydroxyl groups have profound effects on the selectivity of glycosidase inhibition, it is not easy to predict the effects of such changes [Fleet et al., Tetrahedron Lett., 26 3127-3131 (1985)]. For example, 6-episcastanospermine (2) is a glucosidase inhibitor even though the stereochemistry of the four adjacent chiral centres in the piperidine is similar to those in the pyranose form of mannose [Molyneux et al., Arch. Biochem. Biophys., 251, 450-457 (1986)]. Similarly, 1,7a-diepialexine (3), structurally very similar to the powerful mannosidase inhibitor swainsonine (4), is an inhibitor of fungal glucan 1,4-.alpha.-glucosidase [Nash et al., Phytochemistry, submitted for publication]. Also, .beta.-C-methyl deoxymannojirimycin (5) is a strong and specific .alpha.-L-fucosidase inhibitor and has no effect on human liver .alpha.-mannosidase [Fleet et al., Tetrahedron Lett., 30, In Press (1989)]. ##STR1##
With a few exceptions [Raymond and Vogel, Tetrahedron Lett., 30 705-706 (1989)], sugars have been the starting materials used in the synthesis of such compounds as castanospermines [such as (2)], Setoi et al., Tetrahedron Lett., 26 4617-4620 (1985). Hamana et al, J. Org. Chem., 52, 5492-5494 (1987) and Fleet et al., Tetrahedron Lett., 29, 3603-3606 (1988); alexines [such as (3)], Fleet et al., Tetrahedron Lett., 29, 5441-5445 (1988); and homonojirimycins [such as (4)], Anzeveno et al., J. Org. Chem. 54, 2539-2542 (1989). Invariably in the syntheses of these compounds with five adjacent chiral centres and six or seven adjacent functional groups, the strategy chosen has been to start from a hexose and to introduce the additional chiral centre late in the synthesis. An alternative is to start from derivatives of heptoses, that is by very early introduction of the additional chiral centre.
Relatively few studies have been reported on the protecting group chemistry of even readily available heptonolactones [Brimacombe and Tucker, Carbohydr. Res. 2, 341-348 (1966)]. Likewise, only a few examples of syntheses from heptose derivatives have been reported. One neat example is described by Stork et al., J. Am. Chem. Soc. 100, 8272-8273 (1978). Recently, a research group led by co-inventor Fleet herein has found that suitably protected heptonolactones can be powerful and readily manipulatable chiral pool materials. See Bruce et al., Tetrahedron 45, In press 1989, and copending application Ser. No. 07/352,068, filed May 15, 1989 now abandoned.